Western blot - Materials and Methods - Regulation of Actin dynamics by Formin in early Drosophi

Chapter 2. Materials and Methods

2.1 Materials

2.2.5 Western blot

Embryos were staged from 1.5 to 3 hours on apple-juice agar plates and dechorionated in 50% Klorix bleach for 90 seconds. The dechorionated embryos were collected into a Eppendorf tube and weighed. The weight of the embryos was determined (~1mg =100 embryos). The embryo was snap frozen in liquid nitrogen. The embryo were homogenized in 1X Lämmli buffer with the volume to make the final concentration 20 embryos/μl. The sample was heated to 95°C for 5 min and centrifuged at 14,000 rpm for 1 min. The supernatant (protein extracts) corresponding to 10-30 embryos were loaded on the SDS-PAGE. The proteins from the gel were transferred onto a nitrocellulose membrane using a semi-dry transfer for 1 hour at 60 V/gel. The membrane was blocked in 5% milk powder in PBST (fresh made) for at least 30 min and incubated with primary antibody either overnight at 4°C or 2 hours at room temperature. The membrane was rinsed with PBT for three times and 4X15 min PBST washing followed. The membrane was incubated with secondary antibody for 1 hour at room temperature, protected from light. The membrane was rinsed in PBST for three times and washed with PBST for 4X15 min. The bands were detected using the Odyssey CLx Infrared Imaging system.

43 2.2.6 Immunoprecipitation

Protein A beads were washed with PBS. After 1 hour incubation with rabbit DiaC antibody (antisera and purified antibodies) at 4 °C, the beads were washed with PBS for three times and kept on ice. The staged embryos were collected on the apple juice plate, and dechorionated with bleach, then weighted and frozen in liquid nitrogen. The embryos were homogenized in PIPA buffer using Dounce homogenizer. 1 ml RIPA buffer were required for 100mg embryo. The lysate was centrifuged at 14,000 rpm at 4°C for 15 min. The supernatant was added to the antibody-loaded beads as Input and rotated on a wheel for 2 hours at 4°C. The beads were centrifuged with 500 g for 5 min. The supernatant was taken as unbound sample. The beads were washed with RIPA buffer for three times. 2X laemmli buffer was added to the beads and boiled for 5 min. The supernatant was taken after centrifugation at 14,000 rpm for 1 min as bound sample.

According to (1μg=100embryos), the Input, (~10 embryos), unbound (~10 embryos) and bound (~500 embryos) samples were loaded on SDS-PAGE and followed by western blot.

2.2.7 Fractionation of embryos

The dechorionated wild type embryos were homogenized in fractionation buffer using Dounce homogenizer. The lysate was considered as total fractions.

The lysate was centrifuged at 2500 rpm for 5 min at 4°C for two times to precipitate the nuclei. Supernatant was transferred into a new Eppendorf tube and centrifuged at 14,000 rpm for 15 min at 4°C. The lipid layer was removed by an aspirator. The clear supernatant (cytoplasmic fraction) was transferred to a new Eppendorf tube. The pellet (membrane fraction) was washed with fractionation buffer for 3 times. The total, cytoplasm, membrane fraction samples were added 2X laemmli buffer and heated 95 °C for 5 min and loaded on a

Materials and Methods

SDS-PAGE, followed by western blot detection. α-Tubulin was also detected as loading control.

2.2.8 Generation of dia^sy5 and Ced-12^2L367germline clone

The germline clone was performed following the instruction by Chou and Perrimon (Chou and Perrimon, 1992) with minor modifications. The heat shock for inducing flippase was performed at 37°C for 60min per day for two days (24-48 hr and 48-72 hr) after hatching.

2.2.9 Generation of transgenic fly

The transgenic flies were generated with either P-element transposon system or an attB/phi-C31-based integration system (Bischof et al., 2007;

Bownes et al., 1990). The generation process was followed standard protocol (Wenzl et al., 2010) (http://wwwuser.gwdg.de/~jgrossh/method).

2.2.10 Mapping of unknown mutants with meiosis recombination and deficiency

Meiotic recombination mapping was used to narrow down the suspicious mutant region. Frt2L2R{w+}/al dp b pr Frt2L, 2L367 virgins were collected. The heterozygous for the chromosome carrying 2L367 mutant and recessive markers and the Frt2L2R{w+} recombined during meiosis and various classes of recombinant chromosomes went to the female egg and detected by crossing with al dp b pr Bl c px sp/SM1 males. The position of mutation was determined according to the proportion of different recombinant chromosomes. To get the finer localization of mutant, the complement test with deficiencies was done.

For the complement test with deficiencies, the mutant virgins were crossed with the males containing the molecular defined deficiencies on II chromosome.

If the mutant/deficiency progenies are viable and fertile, then the mutant is out of

this deficiency region. If the mutant/deficiency progenies couldn’t be found, i.e.

the deficiency cannot complement the mutation, it means the mutation is located within the region of the deficiency. The deficiency region covers several genes.

In order to know which gene is mutated in the mutant line, the complement test with specific genes which were in the suspicious deficiency region was carried out. The cross strategy is the same as above. If the mutated gene cannot complement the original mutant, it means they are the same gene.

2.2.11 Embryo fixation and immunostaining

The embryo fixation and immunostaining process were followed standard protocol (Wenzl et al., 2010) (http://wwwuser.gwdg.de/~jgrossh/method).

2.2.12 Injection of CK666 and Histone-Alexa488

CK666, Arp2/3 inhibitor, was dissolved in DMSO. WT and dia germline clone embryos were dechorionated, dried in a desiccation chamber for 10 min, covered with halocarbon oil and injected posteriorly with desired concentration of CK666. DMSO was injected as control. After injection, the embryos were incubated for ~10 min and subsequently fixed. The vitelline membrane was removed manually. The embryos were collected in Eppendorf tube, washed by methanol and kept -20°C.

To track the cell cycle, Histone-Alexa488 was injected into the WT and 2L367 germline clone embryos with the final concentration of 2 mg/ml. The preparation of embryos was described above. After covered with halocarbon oil, Histone-Alexa488 was injected posteriorly. The fluorescent movie was recorded at the spinning disc microscope with a 25X oil objective.

Materials and Methods

46 2.2.13 Induction of shibire phenotype

Embryos from shibire heterozygous or homozygous females were collected, kept at 32°C in a water bath for 30 minutes after dechorionation. The embryos were fixed as described previously.

2.2.14 Live imaging

Embryos were dechorionated, lined up, glued on to a coverslip and covered with halocarbon oil. Fluorescent live-images were taken either at the LSM with a 63X oil or glycerol objective or at the spinning disc with a 40X oil objective. Differential interference contrast microscopy movie was recorded at the spinning disc microscope with a 25X oil objective.

2.2.15 Fluorescence recovery after photobleaching (FRAP)

In order to check turnover rate of Dia-GFP on membrane, bleaching of UASp-Dia-GFP under the driven of Maternal GAL4 was carried out in a given area using 100% laser power and 50 iterations at a scan speed of 5.

For examining the membrane property during cellularization, the furrow and furrow canal labeled by GAP43-venus, palmityolated-YFP and 117GFP in wild type and dia germline clone background was bleached. From the surface view a range of Z-stacks were used to track the invaginating furrow canal during cellularization. The 100% laser power and 50 iterations were used for bleaching, and the recording speed was at 5 or 6 depending on how many z-stacks were taken. The other approach was doing FRAP from side view. In this case, the glycerol objective was used. The measurement was done with FIJI.

Chapter 3. Results

3.1 Actin polymerization activity of Dia is suppressed by Cip4

3.1.1 Approaches to identify the potential Dia interactor

The activity of Dia is tightly regulated in eukaryotic cells. The intramolecular interaction between DID and DAD makes Dia in an autoinhibited state in the cytosol (Chesarone et al., 2010). The activation of Dia is achieved by binding of a RhoGTPase to GBD that relieves the autoinhibition via interrupting the interaction between DID and DAD. Meanwhile Dia is recruited to the membrane by RhoGTPase or other factors (Chesarone et al., 2010).

However, in vitro studies showed that RhoGTP in a physiological concentration cannot fully reconstitute the release of Dia autoinhibition (Grosshans et al., 2005; Li and Higgs, 2003), suggesting that additional factors are involved in activating Dia in vivo.

Dia localizes at the membrane, especially is enriched at the furrow canal in the cellularization stage of Drosophila embryo (Figure 3.1). However, by western blotting of fractionation of same stage embryos, I could show that the majority of Dia is in the cytosol; only a small fraction shows up in the embryo membrane extraction (Figure 3.2). The cytoplasmic Dia is considered to be inactive, since Dia is recruited to membrane when it is active (Chesarone et al., 2010).

Results

Figure 3.1 Dia localizes at the membrane. Immunostaining of Dia in cellularization stage of Drosophila embryo. Dia localizes at membrane, and is enriched at the furrow canal.

Figure 3.2 The majority of Dia is in cytosol. Fractionation shows distribution of Dia in the embryo. Only small portion is attached with membrane, while a large amount of Dia is in cytosol.

The absence of α-tubulin in the membrane fraction indicates that the membrane fraction is not contaminated by cytoplasmic fraction. 30 embryos were loaded in each lane.

To identify those unknown factors, I planned to use immuno-precipitation to pull down Dia and the associated protein complex, followed by Mass-Spectrum analysis. The membrane fraction of Dia will be used as a starting material for immuno-precipitation. In our lab we have rabbit and guinea pig source serum against Dia C terminal fragment (termed DiaC in the following text) which works nicely in immunostaining. However, in the western blotting, rabbit source serum shows a stronger background (Figure 3.3 A). On the other hand, DiaC is conserved in the formin family. In an attempt to get a more specific antibody, I used Dia N terminal fragment (termed DiaN in the following text) as the antigen to immunize rabbit and guinea pig. However, no specific bands were detected using DiaN serum both from guinea pig and rabbit (Figure

3.3 A). To remove the background, affinity purification of DiaC rabbit serum was employed (Figure 3.3 B). The background was reduced after affinity purification,

purification,

Figure 3.3 Western blot and immune-precipitation by Dia andtibodies. (A) Dia can be detected by DiaC antibodies raised in guinea pig and rabbit, and guinea pig antibody preforms better in western blot. However, DiaN antibodies couldn’t detect Dia band. (B) After affinity purification of DiaC rabbit serum, the unspecific bands are reduced. (C) Dia can be pulled down with serum and purified antibody. Detected by GP antibody. Empty beads were used as a control.

Results

purification, though there were still some unspecific bands detected.

Endogenous Dia was immuno-precipitated using those antibodies (Figure 3.3C).

The purified antibodies can be used in large scale immune-precipitation and mass spectrometry which will be done in the future.

The other approach for Dia IP is using GFP binder to pull down Dia-GFP complex from Dia-GFP transgenic fly embryos. Five UASp-GFP-Dia lines were generated by Dr. Christian Wenzl in our lab previously (Figure 3.4 and 3.5).

However, the expression level when driven by maternal GAL4 is much higher than endogenous level (Figure 3.5 B). I checked the localization of GFP-Dia using live imaging. Nuclear exclusion of GFP-Dia was observed in these embryos. UASp-GFP-Dias were introduced in dia^sy5, matGal4 flies by crossing.

After inducing the germline clones of dia^sy5 by Flipase-Frt system, the ectopic GFP-Dia can partially rescue dia^sy5 with a rescue rate of ~50%.

Figure 3.4 Schematic representation of GFP-Dia constructs. The GFP with flexible linker was added at C or N terminal of Dia

In order to get a better transgenic fly in terms of expression level and rescue capability, we did another round of making transgenic fly. We reasoned that the GFP at N-terminal could affect the Dia protein folding, resulting in a failure to rescue completely. A flexible linker with the amino acid sequence of AAAGSTGSGSSG was introduced between GFP and Dia. However, the linker did not show any improvement (Figure 3.5).

Figure 3.5 The localization and expression level of GFP-Dia in 10 lines (A) Live images of different GFP-Dia lines. Addition of GFP with linker at C terminal improves the localization of GFP-Dia. The cell border was shown in high magnification. All images were taken with the same settings. (B) Western blot showed that the level of GFP-Dia is much higher than endogenous level. 15 embryos were loaded in each line. Tubulin is detected as loading control.

It was previously found in our lab that N terminal fragment of Dia is responsible for protein localization. Adding extra amino acid at the N terminal may have an effect on the localization function. To overcome this problem, GFP tag was translocated at the C terminal fragment of Dia with the flexible linker.

Meanwhile, a TEV cleavage site was also introduced between the linker and Dia. Four lines were generated after plasmid injection; two lines are with the pUASp-Dia-tev-linker-GFP insertion into the X chromosome and the other two

Results

lines are into the third chromosome. Membrane localization of Dia-GFP could be observed, though there was still nuclei exclusion distribution (Figure 3.5).

In the fixed sample, F-actin intensity in Dia-GFP is higher than wild

type embryo which was stained in the same Eppendorf tube, suggesting the

Figure 3.6 The ectopic Dia-GFP induces F-actin polymerization. (A) Dia-GFP and WT embryos were stained in the same tube, and were distinguished by GFP booster signal. The phalliodin fluorescence intensity is much higher in Dia-GFP embryos than in wild type embryos, indicating the ectopic Dia-GFP can induce F-actin polymerization. (B) Quantitative analysis of phalliodin fluorescence intensity in wild type and Dia-GFP embryo.

type embryo which was stained in the same Eppendorf tube, suggesting the activity of ectopic Dia-GFP in the embryo even though without extra Rho activity (Figure 3.6). However, the rescue rate is not improved (Table 3.1).

Table 3.1 The rescue rate of different transgenic Dia-GFP construct Dia-GFP

construct

da^sy5, matGal4 67;

UASp-GFP-Dia

UASp-DialinkerGFP;

dia^sy5, matGal4 67

UASp-Dia-linker-GFP;

dia^sy5, matGal4 67

Rescue rate

In order to check Dia mobility at the membrane, FRAP analysis was done using Dia-linker-GFP embryos. The signal on the membrane recovered within the range of minute. Compared with other membrane associated proteins, such as Slam and PDZ domain containing protein, Dia showed faster mobility (Acharya et al., 2014).

Figure 3.7 Mobility of Dia is fast. Dia-GFP is enriched at the membrane, as indicated by yellow arrows. The first image was taken before bleach, and the second was taken immediately after bleach. The cytoplasmic signal is hardly bleached, because of the fast exchange in cytoplasm. But the membrane signal completely disappeared after bleaching (the second yellow arrow). The following images were taken every 5 sec, and the signal on the membrane recovered in less than 1 min (the third yellow arrow).

Results

54 3.1.2 Cip4 is an interactor of Dia

S. Bogdan and colleagues (Yan et al., 2013) found Cip4 and Dia can form a complex in S2 cells, which was shown by Co-immuno-precipitation. To confirm this result, binding assay was performed with purified proteins. Dia C terminal half and N terminal half were purified as indicated (Figure 3.8 and 3.9).

Cip4 binds to DiaC with a KD of ~100 nM (Figure 3.10).

Figure 3.8 Schematic representation of proteins purified in this study.

Figure 3.9 Purified proteins used in this study. The samples were loaded on SDS-gel and stained with Coomassie Blue.

Tabel 3.2 Purified proteins in this study

Protein Total amount of

LB culture

Column yield

ZZ-DiaC-Hisx6 3 l HisTrap HP column (1 ml) ~10 mg ZZ-DiaN-Hisx6 3 l HisTrap HP column (1 ml) ~10 mg

Cip4 0.5 l GSTrap HP column (1 ml) ~1 mg

Cip4∆SH3 0.5 l GSTrap HP column (1 ml) ~1 mg

GST-Cip4∆FBAR 0.5 l GSTrap HP column (1 ml) ~1.8 mg GST-Cip4∆FBAR∆SH3 0.5 l GSTrap HP column (1 ml) ~1.1 mg

GST-SH3 0.5 l GSTrap HP column (1 ml) ~1.8 mg

Profilin 1 l poly-L-proline column ~20 mg

Figure 3.10 Physical interaction between Cip4 and Dia. (A) The binding of Dia to Cip4 was detected by binding assay. DiaC, rather than DiaN, could bind to Cip4. (B) Different amount of DiaC were added to GST-Cip4 or GST beads. SDS-Gels were stained with Coomassie Blue.

Results

3.1.3 Cip4 inhibits Dia actin polymerization activity in Pyrene assay (Pyrene assay was done by M. Winterhoff in Prof. J. Faix lab)

Next, we wondered whether the binding of Cip4 show some effect on Dia actin polymerization activity. Pyrene assay was employed to test the actin polymerization activity of Dia. Compared with dDia1 P2 (dictyostelium formin with two poly-proline stretches), ZZ-DiaC showed stronger actin polymerization activity (Figure 3.11 A). In the titration experiment, a series of ZZ-DiaC dilution from 0.125 nM to 1 μM was used. 2.5 nM of ZZ-DiaC was found to be sufficient for polymerizing actin filaments. This is similar to the actin polymerization activity of mDia1 (Li and Higgs, 2003) (Figure 3.11 B, C and D).

Figure 3.11 Dia is a strong actin nucleator shown in Pyrene assay. (A) Dia showed strong actin nucleation activity compared with P2. (B-D) Dia induced actin polymerization at indicated concentrations. 2.5 nM Dia (green line in B) could induce sufficient actin filaments which can be detected by Pyrene assay.

It has been reported that the activity of Dia is inhibited by the intramolecular interaction of DID and DAD domains as mentioned previously (Campellone and Welch, 2010). Theoretically, DiaN inhibits DiaC activity in the ratio of 1:1. However, in the titration inhibition assay, we found 10X more DiaN was needed for the inhibition (Figure 3.12). One possibility is that ZZ-DiaN may be not stable in lower concentration. After dilution, ZZ-DiaN lost the inhibition activity in a few minutes (data not shown).

To test whether Cip4 was able to affect actin assembly, we added increasing amounts of purified Cip4 protein to 10 nM ZZ-DiaC in the actin pyrene assay. We could show that Cip4 inhibited Dia activity in a concentration dependent manner (Figure3.12). 100 nM (10X more than DiaC) of Cip4 is sufficient for inhibition. 200 nM of Cip4 inhibited DiaC activity more efficiently, almost comparable to autoinhibition.

Figure 3.12 Cip4 inhibits Dia actin polymerization activity. Polymerization of actin (2 mM, 10% pyrene-labelled) in the presence or absence of DiaC, DiaN, Cip4, Cip4∆SH3 at the concentrations indicated. Cip4 inhibits Dia actin polymerization in a concentration-dependent manner. However, Cip4∆SH3 couldn’t inhibit Dia activity as effective as by Cip4. Normalized curves are shown.

It was reported that SH3 domain could bind to proline-rich domain and the binding is involved in many cellular process (Aspenström, 2014). S. Bogdan and colleagues (Yan et al., 2013) showed that in S2 cells, the interaction of FH1 domain (proline-rich domain) and SH3 domain is crucial for colocalization of Dia

Results

and Cip4 in the cell periphery. So we tested whether SH3 domain is necessary in the inhibition effect of Cip4. In the pyrene assay, Cip4∆SH3 couldn’t inhibit DiaC activity as efficiently as Cip4 full-length.

Next we checked if SH3 domain itself is sufficient to inhibit Dia activity.

GST-SH3 domain was purified and used in actin pyrene assay. It was shown that GST-SH3 could inhibit DiaC activity. However, this inhibition needs higher molar excess of GST-SH3 (Figure 3.13).

Figure 3.13 GST-SH3 is sufficient for inhibiting Dia activity. GST-SH3 can inhibit actin polymerization activity of Dia, but a relatively high concentration of GST-SH3 is needed.

3.1.4 Cip4 inhibits Dia actin nucleation activity shown by TIRF microscopy (TIRF microscopy assay was done by M. Winterhoff in Prof. J. Faix lab)

Pyrene assay is a bulk polymerization assay, which cannot distinguish the actin nucleation and elongation activity. However, it was reported that Dia has both activities (Campellone and Welch, 2010). In order to test whether the inhibition is due to a reduced nucleation activity, Total Internal Reflection Fluorescence (TIRF) microscopy was used in this study. As shown in Figure 3.14, the single actin filament could be observed using TIRF microscopy and it

Im Dokument Regulation of Actin dynamics by Formin in early Drosophila embryogenesis (Seite 58-0)